With the ability to study gene expression, trace axon pathways, and experimentally analyze axon guidance, cells in the paths of growing axons have been localized, but their identity and fates are not well understood. Our studies have outlined the paths of mouse retinal ganglion cell axons as they diverge to both sides of the brain within the optic chiasm. At this site we uncovered novel cellular specializations consisting of both early neurons and radial glia. In vitro analyses indicate that these cells present clues for axon divergence. In the first aim, in the period before retinal axon growth (E8-11), the early development of these cell populations (form, arrangement, and epitope expression) will be charted through immunocytochemistry, receptor- ligand binding, and in situ hybridization. The birthdates and lineage relationships of these cell will be determined by BrdU and retroviral labeling and their fates (transitions in form, and putative transience) investigated by thymidine labeling in combination with immunolabeling. The reciprocal cell-cell interactions between the early neurons and glia will be studied in dissociated cultures. In the second aim, the patterning of the terrain of the optic chiasm and ventral diencephalon, will be defined by combining in situ hybridization and immunocytochemistry. Analyses of mice deficient of these genes will address to what extent these genes regulate the patterning of the ventral diencephalon. The hypotheses of this work are: that early developing neurons and glia are organized in domains which are important for axon guidance, but are transient in their form, molecular expression and even existence; that expression patterns of regulatory genes in the ventral diencephalon correspond to these cellular domains; that these genes are critical to the development of these cells and ultimately to patterning of the brain. These analyses will interface with Dr. Dodd's studies of induction and early patterning of the anterior of the anterior neural tube, and relate to Dr. Jessell's studies of patterning in the spinal cord, in this poorly understood mid-gestational epoch of brain development. GRANT-P01NS305320003 Understanding the mechanisms that control the identity and diversity of cell types generated in the developing vertebrate nervous system is essential to a complete understanding of brain function and pathology. There is increasing evidence that secreted growth factors of the TGFbeta- like genes (GDF7, BMP6 and BMP7) that are expressed by roof plate cells, a specialized dorsal midline glial cell type, have an essential role in the specification of dorsal neural cell types. The function of GDF7, BMP6 and BMP7 will be assayed by analyzing the phenotype of mice in which each of these genes has been eliminated, singly and in combination, by gene targeting. A panel of markers of dorsal cell types will be used to assess developmental defects in CNS patterning in these mutant mice. These genetic studies will be complemented by in vitro assays of dorsal cell differentiation that are intended to define the potency and selectivity of action of these three proteins. Finally, the signaling properties of the roof plate will be assessed using a molecular genetic approach to the selective ablation of roof plate cells by expression of a toxin gene in transgenic mice. In the long term, these studies should provide important information on the generation and normal function of CNS neurons and the underlying causes of neuronal dysfunction.